A new report shows that extending Medicaid to low-income individuals can increase the number of trips they take to the emergency room — a costly service that many proponents of expanded Medicaid said would be used less often.
The study by Sarah Taubman of the National Bureau of Economic Research and colleagues is the third in a series detailing results from an Oregon-based experiment begun in 2008, in which low-income adults were randomly given Medicaid coverage. The adults were later surveyed to gauge the effects of the coverage on their health and well-being.
Called the Oregon Health Insurance Experiment , it was designed to shed light on the effects of guaranteeing Medicaid coverage for low-income, uninsured adults, an important topic in the United States since insurance for the poor is a big component of the 2010 Affordable Care Act.
While the previous two studies from this group used surveys and in-person interviews to evaluate impacts of the experiment, this latest study , published in the 2 January Science Express, uses emergency room administrative records from nearly 25,000 people for 18 months to get a more detailed look at Medicaid's impacts. The research is free to the public with registration.
Proponents of expanding Medicaid access to the poor have argued that low-income individuals with this new health care access will take advantage of primary care visits and have better health, and thus use the emergency room less. But there has not been much hard data on this topic.
By comparing emergency room administrative records from low-income individuals with and without Medicaid access, Taubman and colleagues discovered that those with insurance made the trip to the ER about 40 percent more often than people with comparable incomes and no insurance, with most of the visits for situations classified as "non-emergent" and "primary care treatable."
Not only did they use the emergency room more, those with the expanded Medicaid coverage didn't show improved health in tests for medical conditions that can be managed with proper care, like hypertension, diabetes and cholesterol. These are conditions that might be expected to show a change within the time interval analyzed in the study.
Thus, the overall picture from the Oregon Health Insurance Experiment suggests that having insurance doesn't necessarily lead to improved health outcomes. Results from previous studies within the experiment suggest that the coverage does seem to give people more peace of mind, which may be related to the fact that it shields them from medical financial woes.
The report's findings may be especially notable as more Americans gain health care coverage with the Affordable Care Act in 2014, writes Columbia University economics researcher Raymond Fisman, in a related Policy Forum  in the journal.
The new study indicates that "we have good reason to anticipate a large increase — and almost surely not a decrease — in traffic to already overburdened emergency departments across the country," Fisman said. "Whether or not you think universal coverage is a good idea, we had better start planning for it."